Published by The Lawfare Institute
in Cooperation With
The Senate Select Committee on Intelligence’s (SSCI’s) groundbreaking study on the CIA’s use of enhanced interrogation techniques (EITs) has been receiving attention again due to the release of the movie “The Report.” “The Report” focuses on the efforts of SSCI investigator Daniel Jones to research and eventually publish the so-called “Torture Report” that was released by the SSCI in December 2014 (referred to here as the “study,” to distinguish it from the movie title). In a companion Lawfare piece, I documented the film’s numerous inaccuracies with respect to how it portrays the CIA’s response to the SSCI investigation, based on my own experience with those events as a lawyer in the CIA’s Office of General Counsel. This post has a different purpose—namely, to document the significant shortcomings in the SSCI study that have little to do with the important debate over whether EITs were legal, moral or effective. These shortcomings have prevented the study from living up to its potential as a historical record and as a vehicle for long-term reforms. To the extent the study is held out as the role model for effective congressional oversight, it’s important to have a serious dialogue about these limitations—so that we can learn from these mistakes and avoid them in the future.
The SSCI study is undoubtedly an important document, and I am not suggesting that it be disregarded in its entirety. In particular, Part II of the study, which is a detailed, mostly factual chronology of the creation and operation of the interrogation program, is by far the most comprehensive public accounting available. It will be a valuable resource for years to come. In addition, if the study ultimately contributes to its goal of ensuring that the U.S. government never repeats these mistakes, it will have generated a critical achievement despite these flaws.
Notwithstanding these attributes, the SSCI study is still a limited document that didn’t live up to its full potential. The release of “The Report,” along with the five-year anniversary of the study’s publication, provides an opportunity to reflect on its long-term impact. In looking back, I believe that if the study had avoided the mistakes set forth below, it would have been more respected within the government and more persuasive with the general public. Instead, these problems allowed the study to be easily rejected by those who took issue with any one of the problems identified below. Most CIA officers I worked with, particularly the generation that joined the agency after the detention and interrogation program had been shut down, recognized that there were significant problems with the program. I knew few officers who were ardent defenders of the use of EITs (with the benefit of hindsight), and no officers who wanted to return to that practice. But virtually everyone I knew in the CIA and elsewhere in government immediately saw the SSCI study as a biased, agenda-driven document because of these flaws, and, as a result, they found little utility in it. They believed that the aim of the study was not just to ensure that EITs were never used again but also to inflict long-term institutional damage to the CIA and its officers. In addition, because of its misguided focus on “whether torture works,” the study ensured that the public discussion would focus on this issue and not any of the additional, more complicated questions the program raised relating to morality, legality, oversight, accountability and shared responsibility.
As documented in my companion Lawfare piece, I had no involvement in the detention and interrogation program (I joined the CIA in 2010 and left in 2018), and I am not a defender of the use of EITs. I strongly supported the large-scale declassification of information relating to the program (an effort I helped coordinate) and the need to have a public accounting of what occurred. In addition, I believe that strong, effective oversight is essential for the proper functioning of the CIA and for maintaining its credibility within the government and the general public. Based on these views, I believed that the SSCI study (if done right) presented a unique opportunity not only to help ensure that EITs were never deployed again but also to provide valuable insights into how the government should manage and oversee sensitive covert action programs in a time of crisis. It failed to live up to this potential in this regard.
Many of the study’s shortcomings documented below reflect deliberate choices made by the SSCI investigators, led by Jones. It is true that they had limited time and resources, and that every choice in an investigation requires making trade-offs. But Jones and his colleagues also spent four years generating a more-than-6,000-page report, so they cannot reasonably argue that any of these issues were beyond their reach. By presenting a more limited picture in the study, it was easier for Jones to support his thesis (which is also reflected in “The Report”) that the CIA cannot be trusted in any circumstance. “EITs never worked, and the CIA lied to everyone about everything” is a fairly simple argument. The trouble is that the real story of the interrogation program’s failures is much more complicated, and the study failed to grapple with the difficult questions presented by this more nuanced history.
In documenting the study’s shortcomings, my primary intent is not to provide substantive counterarguments to its factual findings and analytical assessments. Readers should refer to the CIA’s official response to the study for that and make their own judgment. Rather, my aim is to outline the study’s strategic mistakes, which have limited its long-term impact. In other words, this is not a rebuttal, but a road map for how the study could have been much better. Those strategic mistakes include the following:
First, the study focused too much on whether EITs were effective and the CIA’s alleged misrepresentations relating thereto. I won’t belabor this point and will instead refer readers to David Cole’s excellent article about the study, where he observed that “by focusing on whether the program worked and whether the C.I.A. lied, the report was critically misleading.” As Cole explained, this focus distracted from a discussion over legality, morality and shared responsibility. “So why did the committee focus on efficacy and misrepresentation, rather than on the program’s fundamental illegality,” Cole asked. “Possibly because that meant it could cast the CIA as solely responsible, a rogue agency.”
The study’s singular focus on this issue was also a mistake because it was not completely convincing on the merits. Cole, a leading human rights advocate who is a strong critic of the CIA, carefully examined both the study and the CIA’s response and came away convinced that the CIA was “treated unfairly” by the SSCI study in this regard. If the study had argued that the CIA had sometimes exaggerated its claims relating to EITs, that EITs were counterproductive in addition to being illegal, and that the agency did not sufficiently attempt to utilize noncoercive techniques, that would have been one thing. But the study’s much more dramatic claim—that EITs produced no valuable intelligence, and that the agency lied about every single claim of success—was simply not credible, even for an ardent critic like Cole. As Cole noted, this claim also had little bearing on whether these techniques were legal or moral, and it also excused the responsibility of other institutions like the White House and Department of Justice.
From my perspective, testing the credibility of Jones’s central claim can be accomplished without reexamining the substantive arguments on both sides of the debate. As I explained in my companion post, evaluating a conspiratorial allegation like this can be done in some cases not by examining it on the merits, but by asking what discrete acts would have been required to maintain the alleged conspiracy, and are the perpetrators’ actions consistent with that? Here, the conspiracy alleged by Jones—that the CIA consistently and deliberately lied about every claim relating to the interrogation program for a 12-year time period (2002-2014)—would have necessarily required a well-coordinated, secret pact that reached all corners of the agency. Everyone who ever touched information relating to the program would have needed to be in on it. There would have been countless emails, instant messages, edits to memos, and the like documenting the conspiracy—constantly reminding officers of the party-line on “X claim” and the need to consistently support it. Yet Jones and his team located not a single record among the agency’s 6.3-million-page documented production that referred to this far-flung conspiracy. It is true that Jones found information that he believed refuted or undermined certain claims that the agency made, and that there were internal detractors of the program. But Jones did not uncover any supporting evidence that the agency was, in real time, engaged in a concerted campaign to lie to everyone about every success claim relating to the program.
Agreeing with Jones’s assertion that the CIA consistently and deliberately lied about the intelligence obtained from EITs also necessarily requires one to believe a much more serious claim: that the agency’s counterterrorism officers were actively engaged in a disinformation campaign that would have harmed ongoing operations and threatened national security. Deliberately lying about the intelligence gained from the interrogation program would have simultaneously required (a) a concerted effort to maintain this scheme, a laborious endeavor that would have detracted from ongoing efforts to prevent additional attacks and (b) creating false intelligence reports and assessments, which would have been relied upon by others in the government, who would then be chasing false leads. It is both insulting and nonsensical to suggest that these officers, who dedicated their lives to preventing another terrorist attack, would have done so.
Those who worked in government immediately understood that Jones’s allegations were not plausible for the reasons I’ve outlined, separate and apart from the arguments on the merits put forward by the CIA. None of this is to say that the study should have conceded that “torture works” or that all of the agency’s claims about intelligence obtained from the program were accurate. But the study would have been much more credible and effective if it had instead presented a more nuanced argument—that CIA did not properly vet and continuously reevaluate all of its success claims, that on some occasions it exaggerated or misstated them (as conceded by the CIA in its response to the study), that it should have not have engaged in a public relations campaign relating to the program, and that it may have been able to obtain some of this intelligence by other means. It also failed to seriously explore to what extent these mistakes were made in good faith, which is a critical point. The study’s more aggressive argument was easily refuted, failed to resonate with the public, and sucked up all the oxygen in the public debate, while distracting from other important issues outlined by Cole in his article and below.
Second, the study did not seriously attempt to provide policy recommendations or proposed reforms. When the SSCI approved the full study and sent it to the CIA in early 2014, it did not contain a single recommendation among its 6,000-plus pages of text. The CIA personnel who worked on the agency’s response to the study were shocked. Four years of painstaking work but no recommendations? Then-Chairman Sen. Dianne Feinstein eventually generated a list of recommendations three weeks after the study was publicly released, but these came only after the SSCI received the CIA’s response to the study, which contained its own more detailed set of proposed reforms (many of which had already been implemented). Feinstein’s recommendations were not particularly insightful and received little attention since they were not released with the study. In addition, there is a significant difference between conducting an investigation with an eye toward reforms (where an investigator would look for patterns and gather evidence that would support certain reforms) versus tacking on a set of cursory recommendations after-the-fact.
There is no problem with the fact that the SSCI decided to make the study mostly a historical examination of the interrogation program; documenting the full story of the program is extremely important. But it would not have been difficult for the committee to conduct its historical investigation with an eye toward forward-looking reforms. The CIA is going to continue to conduct sensitive covert action programs at the direction of the president. Beyond the need to avoid EITs, what lessons can we learn from the interrogation program that will help the agency, the White House and Congress avoid repeating these broader mistakes in the future, especially in times of crisis? The study provided no answers.
Other recent investigations into the CIA have not made this mistake. The Silberman-Robb Commission report into the agency’s intelligence failures relating to Iraq’s weapons of mass destruction capabilities contained more than 200 pages devoted to reforms in its 500-plus-page report. The 9/11 Commission had 60 pages dedicated to reforms in its landmark study of similar length. Even the hyperpartisan Benghazi Committee offered policy recommendations in its otherwise political report.
One reason the SSCI may have chosen to avoid policy recommendations is that many of the more obvious reforms had already been achieved by the time the study was completed. Some were accomplished through legislation (such as the McCain-Feinstein amendment and covert action notification reforms) and executive order, others through internal reforms already implemented by the CIA, and others through a fundamental change in the CIA’s approach to congressional oversight implemented by CIA Director Leon Panetta and the Obama White House. A discussion of these issues, particularly the shift in the CIA’s engagement with Congress, would have required acknowledgments that didn’t quite fit with the narrative that the CIA was incapable of learning from its mistakes or submitting to civilian oversight.
As documented in my companion Lawfare piece, the CIA took a fundamentally different approach to congressional oversight during the Obama years, particularly in the counterterrorism arena—which I’m sure comes as a shock to most readers. SSCI oversight staffers knew as much, if not more, about the agency’s counterterrorism operations than I knew when I was serving as the chief of staff to the CIA general counsel. In fact, it got to the point where the CIA’s counterterrorism briefers were sometimes sharing too much information with the SSCI, often telling the committee about internal executive branch policy deliberations well before issues had been decided. Jones’s SSCI colleagues who worked with the CIA during that time would attest to this.
There are a significant number of forward-looking issues that the study could have examined that were not already addressed by the reforms that had been implemented when the study was completed. Here are just a few:
- Since the SSCI study admitted that the CIA obtained valuable intelligence from detainees, just not as a result of EITs, how was the CIA best able to elicit this information? What noncoercive methods worked best?
- Even if EITs are banned, should the CIA be in the detention and interrogation business at all, even on a much more limited basis? Under what circumstances?
- How should the U.S. government approach situations in which the CIA’s foreign partners are asked to engage in risky activities that could have significant foreign policy implications, like the maintenance of black sites? What role should the secretary of state play in such decisions, and how should it be institutionalized?
- How should the oversight committees respond the next time an administration refuses to provide sufficiently detailed briefings about sensitive covert action programs that it claims are vital to keeping the country safe?
- How should the government balance secrecy against transparency in the covert action arena? Should the government rethink the rules on covert action so that the CIA can conduct traditionally “covert” activities in a more overt and transparent way?
The CIA and the rest of the national security community, which are fundamentally forward-looking institutions, would have greatly benefited from serious dialogue on these issues and many others.
Third, the study did not examine what went right during the course of the detention and interrogation program. An investigation that was intended to be constructive and comprehensive would not only look at the CIA’s mistakes and failures during the program; it would also look at whether there were any successes. Understanding what went right provides important context if you’re trying to completely understand what went wrong and how such mistakes can be avoided in the future.
It is not possible to dismiss this criticism by noting that the CIA did nothing right in the wake of Sept. 11. Consider the following:
- Within three years after Sept. 11, the CIA successfully captured and detained most of the core planners of the largest act of terrorism in American history, many of whom were hiding in some of the most hostile corners of the world. And while Osama Bin Laden and Ayman al-Zawahiri remained at large for the duration of the program, their power and operational influence was greatly diminished due to the unrelenting pressure that the CIA applied to the al-Qaeda network.
- During the same time period, the CIA captured and detained more than 100 additional terrorists outside of this core group. While many of these terrorists were ultimately not deemed to be valuable enough to remain in the CIA program (and some were wrongly detained), the majority were still dangerous, and their capture neutralized the significant threat that they posed to the United States and its allies.
- The CIA captured all of these terrorists; it did not kill them on the battlefield. Going through the extra effort, and associated danger, of capturing these detainees provided the opportunity to gain valuable intelligence from these individuals. And it allowed for the possibility that they could eventually be brought to justice in an American courtroom, although that has proved illusory, unfortunately. This record stands in contrast to the Obama administration’s practice of carrying out targeted lethal action against high-value terrorists, thereby depriving the government of the opportunity to obtain such information (although it is true that capturing these terrorists was not always feasible in places like Yemen or the Afghanistan-Pakistan border region).
- While not stated explicitly, the SSCI study concedes that the CIA obtained valuable intelligence from these detainees that helped prevent additional attacks. The study asserts that none of this valuable intelligence was obtained through the use of EITs, but in making this argument, the study tacitly admits that the CIA successfully obtained much of this information through other aspects of the program.
The CIA did all of this in just a few years even though on Sept. 12, 2001, it had no apparatus in place to capture, detain and interrogate terrorists, beyond its limited experience in carrying out a small number of renditions during the Clinton administration. The agency’s success was nothing short of remarkable, but it’s been overshadowed (understandably) by its use of EITs. The question is: Should we ignore these successes, or try to learn from them? The SSCI study authors chose the former.
Among other things, examining the successes would have been an important part of any “lessons learned” analysis that informed forward-looking recommendations. Comparing what worked with what didn’t work would provide an important contrast that would assist greatly with identifying potential reforms. For instance, the study could have examined how the CIA was able to build such a massive program so quickly in the immediate aftermath of Sept. 11. Anyone who has worked in national security can tell you that one of the things that makes the CIA stand out is how nimble it is and how it can break through bureaucratic logjams that plague agencies like the FBI and the Department of Defense. Another issue the study could have explored is the importance of maintaining the agency’s liaison relationships in far-flung places of the world, even when they may not seem to have strategic importance at the time. In the 1990s, not many people thought that maintaining close ties to obscure Afghan tribal leaders was worth the time and effort, but a few officers in the CIA did, and on Sept. 12, 2001, those relationships were one of the most important resources the U.S. government had at its disposal.
In addition to helping inform potential reforms, including at least some exploration of what went right with the program also would have built credibility with employees at the CIA and other intelligence community agencies. In the wake of Sept. 11, thousands of CIA officer worked tirelessly to prevent another attack, and they still do so to this day. They risked their lives, disappointed their children and ruined their marriages—all to ensure that other American families remained safe. These officers are tough and used to being scapegoated, and they don’t need or expect a “thank you” from their oversight committees. But they do value honesty and objectivity, and the study’s effort to paint the agency’s actions in the wake of Sept. 11 as one giant shit show (for lack of a better term), while failing to provide even a cursory acknowledgment of these successes, revealed a fundamental bias that caused these officers to dismiss the study in its entirety as a biased, agenda-driven venture.
Fourth, the study did not examine how Congress could have done a better job of exercising its oversight responsibilities over the program. The SSCI study contains not a word of critical examination on whether Congress could have done more to prevent some of the abuses during the detention and interrogation program. It lays 100 percent of this failure at the feet of the CIA. I am not going to engage in the debate over who in Congress was told what and when—that’s pointless. But even if one concedes the SSCI’s point that congressional leaders who were briefed on the program were both under- and ill-informed, the uncontroverted fact remains that these leaders were aware that the CIA was operating secret detention facilities overseas and interrogating terrorists at those facilities, and they did not use their powers to demand more information. Yes, they did write a few letters and voice concerns at times, but those limited efforts can hardly be described as robust, effective oversight.
While these congressional leaders are certainly at the bottom of the list when apportioning responsibility for this episode, they cannot be excused entirely for their failure to ask the right questions and to insist on getting those questions answered. They had significant powers at their disposal and failed to exercise them. Unfortunately, the questions of whether they could have done anything more, and whether any reforms could be instituted to prevent this oversight failure from recurring in the future, were unexamined in the study.
Fifth, the study focused too much on career officers, not on agency leadership. Providing a mechanism for accountability was an important purpose of the SSCI study, but the SSCI made odd choices in this regard. Strangely, the study focused much more on rank-and-file officers who carried out the program than the agency’s leadership that oversaw it. This focus played out during the negotiations over the redactions, where the most significant dispute between the agency and the SSCI staffers surrounded whether the pseudonyms that the SSCI had included for these officers would be unredacted (most of the drama over redactions that was portrayed in “The Report” related to this issue). The CIA was rightly opposed to release of the pseudonyms, as it would have taken just a few hours for a “key” to be generated with the true identities of these officers (many of which had leaked in the past), putting them and their families at risk.
While these career officers bore the brunt of the study’s criticisms, the CIA’s leadership was largely unscathed. Remarkably, George Tenet is mentioned only 24 times in the study’s 500-plus pages, even though he was the director when the program was designed and implemented and when the harshest EITs were carried out. His deputy during this entire time, John McLaughlin, is mentioned only three times. Waterboarding was carried out entirely on their watch. They were responsible, along with the White House, for determining what Congress was told about the program. Yet they are barely mentioned in the study, and when they are, they are spared criticism. I know both men and like them very much. But they would be the first to say: The buck stops with them, not the rank-and-file officers.
The disproportionate focus on these career officers does serve the purpose of undermining trust in the CIA institutionally, however. By the time the study was released in 2014, Tenet was long gone, but many of these officers remained with the agency. Focusing on them emphasized that the CIA can never be trusted, no matter who is in charge.
Sixth, the study did not seriously examine the complicity of other executive branch officials. The study focused almost exclusively on the actions of the CIA during this time period, and it largely gives a pass to the White House, National Security Council and Department of Justice, even though they were ultimately in charge of approving and overseeing the program and opining on its legality. If anything, the study excuses these officials’ actions by focusing on the CIA allegedly hiding information from them or misrepresenting them—even though after the study was released, none of these officials complained that they had been materially misled about what was going on.
The lack of attention given to White House and Justice Department officials serves the thesis that the CIA was operating as a rogue agency, but it does so at the expense of the truth. Nor can Jones dismiss this point by arguing that the White House fell outside the SSCI’s jurisdiction. I know of no oversight staffer who would concede this point, and would note that the House Intelligence Committee under Adam Schiff has spent a large portion of its time focused on the actions of the Trump White House, not just agencies under its jurisdiction. The SSCI could have done the same in the study and chose not to.
Finally, the SSCI did not interview CIA officers, who could have provided critical context for the events examined in the study and valuable insights on reforms. The SSCI study is probably the only major congressional investigation in the modern era that did not conduct interviews. Imagine if the Benghazi Committee hadn’t conducted any interviews, or if the 9/11 Commission had done the same. One cannot credibly contend (as Jones has) that interviews were unnecessary, and that the study gets everything right without the benefit of talking to the officers who lived through these events. This is like arguing that someone could read your text messages and emails and have a perfect understanding of what is going on in your life. To state the obvious, these officers could have provided necessary explanations of what certain documents meant as well as valuable context that wasn’t reflected in the documents. They also could have used their experience to provide suggestions on potential reforms.
It is true that the CIA objected to the SSCI interviewing employees about the interrogation program once Special Prosecutor John Durham’s criminal investigation launched, but the agency could not prevent the SSCI from interviewing former employees, and there were quite a few former employees out there who were involved in the interrogation program. Some had even written books about it. Why not talk to them? In addition, the Durham investigation into the interrogation program largely ended in June 2011. The SSCI could have elected to start interviewing employees at this time; the study was not completed until 18 months later, in December 2012. It deliberately chose not to do so, thereby limiting the accuracy of the study. Finally, if the SSCI had pushed hard enough, it likely could have overcome the agency’s initial objections to interviews. Absent invoking the Fifth Amendment, there is no privilege that allows government employees to avoid such interviews due to the pendency of a criminal investigation. During the Benghazi congressional investigation, there were two pending criminal investigations (one into the attack, another into Hillary Clinton’s emails), and dozens of interviews were conducted with executive branch employees. If the political will were there, interviews could have been conducted by the SSCI.
The SSCI study should be applauded for shining a light on the full history of the interrogation program (including its brutality), and for contributing to an effort to ensure that EITs are not used again. But its focus on whether EITs worked almost seems to accept that this issue should be evaluated on a utilitarian basis, rather than on principled legal and moral grounds. This was a mistake, as was the study’s attempt to pin the program’s problems primarily on the CIA, instead of examining this episode as a whole-of-government problem. By ducking these more complicated questions, the study failed to generate a serious dialogue on how we avoid these problems the next time we face a national security crisis.
All statements of fact, opinion, or analysis expressed are those of the author and do not reflect the official positions or views of the Central Intelligence Agency (CIA) or any other U.S. Government agency. Nothing in the contents should be construed as asserting or implying U.S. Government authentication of information or CIA endorsement of the author’s views. This material has been reviewed by the CIA to prevent the disclosure of classified information.